CROSSING THE LINE IN CROSS-SEX FRIENDSHIPS: EFFECTS OF PAST SEXUAL CONTACT AND DATING STATUS ON RELATIONAL MAINTENANCE

Transcription

1 CROSSING THE LINE IN CROSS-SEX FRIENDSHIPS: EFFECTS OF PAST SEXUAL CONTACT AND DATING STATUS ON RELATIONAL MAINTENANCE by MELISSA CHRISTINE EMMETT B.S. University of Miami, 2006 A thesis submitted in partial fulfillment of the requirements for the degree of Masters of Arts in the Nicholson School of Communication in the College of Sciences at the University of Central Florida Orlando, Florida Spring Term 2008

2 2008 Melissa Emmett ii

3 ABSTRACT Given that male-female friendships lack institutional guidelines or scripts to guide everyday interactions, sustaining these relationships can be complicated. Although researchers have recently begun to study relational maintenance patterns in cross-sex friendships, no study has yet to investigate sex s and dating status s direct influences on friendship maintenance. This investigation seeks to add to the field s academic understanding of these personal relationships. Specifically, the researcher sought to uncover the associations between 10 friendship maintenance behaviors, past sexual contact, valence of past sexual contact, and cross-sex friends dating statuses. Quantitative data were collected from 277 participants at a large Southeastern university using a survey methodology. Overall, results indicate that patterns of relational maintenance do differ as a function of friendship level, past sexual involvement, valence of past sexual involvement, and cross-sex friends dating statuses. Past sexual involvement was associated with higher levels of relationship talk, avoidance of negativity, and flirting. In addition, valence of past sexual activity varies across friendship level and associates with higher levels of positivity and flirting. Lastly, dating status significantly affects 3 of the 10 maintenance behaviors: routine contact and activity, flirting, and humor. Linear, quadratic, and cubic relationships were found for these variables, respectively. Implications of results, limitations, and recommendations for future research are discussed. iii

4 ` ACKNOWLEDGEMENTS HWW: Quite simply, words are inadequate to express my appreciation for your contagious enthusiasm, unwavering commitment, and never ending patience. Whether it was a humorous story to encourage or a well-needed push in the right direction to motivate, you kept me going. You kept me interested. Undoubtedly, my most rewarding experiences as a grad student all have one common thread: your influence. Thank you for helping this M.A. in progress, complete the degree! Dr. Katt & Dr. Hastings: I am a better researcher because of the feedback you offered and the advice you shared. Thank you for your valuable time and input. Kirsten & Steve: From beginning to end, the kindness and support you both have offered has not gone unnoticed in this enduring process. The jokes weren t bad either! You are appreciated more than you may know. Gina, My Weekly Date: Grad school isn t always easy, but it s more enjoyable when someone gets what you re doing and what you re all about. You were that person for me. I hope you don t just perceive our experiences to be meaningful ---we ve socially constructed it in our reality! Valerie: Our weekly conversations have always served to put a smile on a face no matter what was going on in the classroom or in my personal life. You are twintastic. Quite simply, thank you for being you. My Family: Before I ever spoke with professors or peers about research, you were among the firsts to hear my thoughts--thank you for continuing to listen as my ideas went through a much needed refinement. Any time you showed a spark of interest in my work this, THIS -- is when I became most proud to complete it. M: Whether you were my much needed distraction or my voice of motivation, you always supported me. Thank you for being you so I could be more me. iv

5 TABLE OF CONTENTS LIST OF FIGURES... vi LIST OF TABLES... vii CHAPTER 1: INTRODUCTION... 1 CHAPTER 2: LITERATURE REVIEW... 4 The Context of Cross-Sex Friendships... 4 Relational Maintenance... 6 Overview of Romantic Maintenance... 6 Friendship Maintenance... 9 Attraction in Cross-Sex Friendships Sex in Cross-Sex Friendships Dating Status CHAPTER 3: METHODS Participants Procedures Instrumentation Relationship Maintenance Behaviors Sexual/romantic history Dating status Friendship level CHAPTER 4: RESULTS Hypothesis One Research Question One Hypothesis Two CHAPTER 5: DISCUSSION Limitations and Future Research APPENDIX A: IRB APPROVAL LETTER APPENDIX B: IRB APPROVED VERBAL CONSENT DOCUMENT APPENDIX C: IRB APPROVED SURVEY REFERENCES v

6 LIST OF FIGURES Figure 1: Linear Relationship of Routine Contact and Activity Across Dating Statuses Figure 2: Quadratic Relationship of Flirting Across Dating Statuses Figure 3: Cubic Relationship of Humor Across Dating Statuses vi

7 LIST OF TABLES Table 1: Primary Factor Loadings for Relational Maintenance Subscales Table 2: Descriptive Statistics for Variables in the Analysis Table 3: MANCOVA Table Table 4: Parameter Estimates Table 5: Means and Standard Deviations Across Levels of Dating Status Table 6: Step Down F Table Table 7: Multiple Regression Table for Valence of Past Sexual Encounter With Friend vii

8 CHAPTER 1: INTRODUCTION Considering that friendships play a key role in people s everyday lives, further understanding of the communication patterns within these relationships will help a communication scholar, and the average person, better understand the interactions, dynamics, and outcomes within this relational context. One particular kind of friendship, an opposite-sex friendship, offers its members an exclusive benefit and is becoming increasingly common in today s society. As Bleske-Receck and Buss (2000) propose, cross-sex friendships are unique opportunities for men and women to learn inside information about the opposite-sex, including information about what members of the opposite sex desire in a mate and how to attract them. In addition, an early study by Booth and Hess (1974) found that more than 35% of men and 24% of women reported a person of the opposite sex as their closest friend. With personal and professional interactions between men and women steadily rising, individuals are presented with a greater number of opportunities to develop meaningful, long-lasting female-male friendships (Monsour, 1996; Sapadin, 1988). Despite this change in the social landscape, literature on the topic remains relatively small compared to other types of relationships. Recognizing that friendships develop within a variety of social settings (professional, academic, personal, etc.), further understanding of how these relationships function is warranted. Even though cross-sex friendships comprise intriguing, beneficial relationships (Griffin & Sparks, 1990; Hays, 1985; Monsour, 1992), they may be particularly fragile and difficult to maintain (Werking, 1997). Although some uncertainty is always present within relationships (Baxter & Montgomery, 1996; Honeycutt, 1993), individuals within cross-sex friendships are likely to 1

9 experience relational uncertainty for a variety of reasons (Afifi & Reichert, 1996; Ficara & Mongeau, 2000). While same-sex friends may experience some of these same problems, other issues arise specifically within friendships between women and men. Research indicates that general rules are cross-culturally recognized for same-sex friendships (Argyle & Henderson, 1984), however, norms for cross-sex friendships remain unclear (O Meara, 1989). In addition, ambiguity exists about the potential romantic or sexual nature of the relationship (Afifi & Burgoon, 1998), and this ambiguity could inhibit the use of pro-social maintenance behaviors (Dainton, 2003). For example, a cross-sex friend interested in a strictly platonic relationship may reduce the use of Positivity and Flirting if he/she feels his/her friend wants to escalate the friendship to a romantic relationship. While past research shows that not all of these relationships involve sexual tension (Monsour, Beard, Harris, & Kurzweil, 1994), Kaplan and Keys (1997) found some degree of sexual attraction, by at least one member of the friendship, underlies many male-female friendships. In fact, ambiguity about the potential romantic/sexual nature of these friendships typically complicates relational functioning (Afifi & Burgoon, 1998). Although researchers debate the impact that sex and attraction exert on male-female friendships, Schneider and Kenny (2000) suggest that sexual desire and attraction play important roles in developing opposite-sex friendships One way opposite-sex friends negotiate the proper content and conduct for their relationship is via routine and strategic relational maintenance. The vast majority of previous research on maintenance processes in interpersonal relationships chiefly focuses on romantic relationships, marriages, and general friendships. Comparatively less research, however, investigates how opposite-sex friendships vary in terms of maintenance behaviors (Messman, Canary, & Hause, 2000). Given that maintenance behaviors function as a way to sustain the status quo of a relationship 2

10 in accordance with a person s goals (Dindia & Canary, 1993) and romantic interpersonal goals are influenced, at least in part, by sexual/romantic attraction, communication scholars benefit from learning more about these associations within a specific type of relationship latent with romantic/sexual uncertainty. Although researchers have recently begun to study relational maintenance patterns in cross-sex friendships, no study has yet to investigate sex s and dating status s direct influences on friendship maintenance. This study will explore how patterns of relational maintenance change as cross-sex friends cross the line with each other and with outsiders. 3

11 CHAPTER 2: LITERATURE REVIEW The Context of Cross-Sex Friendships When trying to define what a cross-sex friendship is, it is often easier to compare and contrast its relational characteristics to other relationship types. As outlined in Rawlins (1982), cross-sex friendships and romantic love relationships overlap in many ways, complicating the task of identifying relational characteristics specific to this particular type of friendship. To start, both relationships are voluntary, personal, and privately negotiated between particular individuals. Additionally, men and women must invest a considerable amount of emotional energy in order to sustain both of these personal relationships. Further, the individual s unique qualities qualify him or her as a candidate for intimacy. According to Greeley (1970), romantic relationships and cross-sex friendships are similar since both extend the open trusting commitment of oneself to another human being (p. 35). Although both types of relationships are founded on similar ground, the communication patterns in these relationships will systematically differ according the type of emotional bond they share with one another. The existing literature assumes heterosexuality in a cross-sex friendship. Rawlins conceptually distinguished between the types of emotional bonds that men and women can experience. These emotional bonds include: friendship (a voluntary, mutual, personal, and affectionate relationship devoid of expressed sexuality), platonic love (a deeply emotional, exclusive, non-sexual relationship, friendship love (an unstable relationship characterized by fluctuating expressions of emotions and sexual affection), physical love (a relationship high in sexual involvement with little emotional commitment), and romantic love (an exclusive, sexual, emotional 4

12 relationship). Since these emotional bonds are not easily identifiable or distinguishable and have the potential to change over time (Rawlins, 1982), many problems often arise between friends of the opposite sex. Although similarities exist between same-sex friendships and cross-sex friendships, some major differences exist, presenting unique challenges to friends of the other sex. Cross-sex friendships must resolve four challenges that O Meara (1989) identifies. The first challenge is that they must successfully develop a shared definition of the type of love (emotional bond) they experience. The second challenge stems from the fact a friendship between a heterosexual man and woman introduces the potential for sexual attraction in the relationship. The third challenge arises because one of the defining characteristics of friendship is equality (Suttles, 1970; Paine, 1974), yet cross-sex friends interactions all occur within a male-dominated society. The last challenge identified by O Meara occurs because friends must use strategic communication to show relevant audiences they do not have a hidden sexual agenda. According to O Meara, all of these challenges continuously vary throughout the lifespan of the friendship: they are never fully resolved, simply fluctuating between being dormant and emergent as situations change. All of these challenges have the potential to create relational uncertainty and cross-sex friends can manage these challenges by enacting in relational maintenance behaviors. Although O Meara s (1989) work highlights that attraction potentially complicates relational functioning by increasing conflict and uncertainty in cross-sex friendships, only a handful of communication scholars actually have examined attraction in male-female friendships. One scholar s work, Reeder (2000), notably stands out from the body of existing research on the topic. Her study found that cross-sex friends experience attraction differently than romantic partners and that other 5

13 types of attraction present themselves within these friendships. Reeder identifies four distinct types of attraction found in cross-sex friendships: subjective physical/sexual attraction (feeling physically or sexually attracted to the other person), objective physical/sexual attraction (thinking the other person is attractive in general, but not to oneself), romantic attraction (wanting to turn the friendship into a romantic relationship), and friendship attraction (feeling close and connected as friends). Experiences of attraction differ in friendships because attraction types can vary in degree, exist separately or together, be symmetrical or asymmetrical, and remain constant or vary over time (Reeder, 2000). To elaborate, these attraction types are not permanent or mutually exclusive. Friends do not always experience the same type of attraction for their friend throughout the relationship s duration: a person can experience one type of attraction towards his/her friend at one point of the relationship and experience another type of attraction towards his/her friend at a different point of the relationship. In addition, multiple types of attraction can be simultaneously experienced. Attraction is defined as symmetrical if both friends experience the same type(s) of attraction towards each other; attraction is asymmetrical if one friend s report of attraction type is not consistent with the other friend s report of attraction type. For all of these reasons, varied experiences of attraction manifest in cross-sex friendships. As a result, friends of different sexes are likely to experience a heightened level of relational uncertainty, making relational maintenance behaviors an even more important part of their communication. Relational Maintenance Overview of Romantic Maintenance Relational maintenance is defined as behaviors that sustain preferred relational characteristics (Dindia & Canary, 1993). Duck (1986) distinguishes between two types of maintenance behaviors: 6

14 carefully selected strategies and everyday, routine interactions. According to Dindia (1991), individuals consciously use strategic maintenance to intentionally obtain a specific goal or result. Routine maintenance behaviors, however, require little, if any, conscious thought to perform and are not usually performed with the expressed goal of maintaining the relationship (Dainton & Stafford, 1993). Regardless of the level of consciousness, performance of both strategic and routine behaviors serves maintenance functions and influence relationship stability and satisfaction (Canary, Stafford, & Semic, 2002). In this study, the term maintenance behavior encompasses both strategic and routine interactions. Early research by Stafford and Canary (1991) delineated five categories of maintenance behaviors in romantic relationships: positivity (i.e., behaviors that create positive and enjoyable interactions), openness (i.e., behaviors that reveal thoughts and feelings about each other or the relationship), assurances (i.e., behaviors that imply commitment to the relationship), social networking (i.e., spending time with friends and family), and task sharing (i.e., performing one s share of the chores or helping with routine tasks). Advice (sharing one s opinions or offering help in solving each other s problems) and conflict management (i.e., behaviors aimed at amicable dispute resolution) were subsequently added to the list of relationship maintenance behaviors (Stafford, Dainton, & Haas, 2000). Initial research on maintenance in romantic relationships established that relational maintenance behaviors moderately and positively correlate with each other. In general, the more an individual performs one maintenance behavior, the more that individual will also perform other maintenance behaviors (Stafford & Canary, 1991; Stafford et al., 2000). Maintenance behaviors not only correlate within individuals, they also correlate within dyads: one partner s use of a maintenance strategy predicts the other partner s use of the same strategy (Dainton & Stafford, 7

15 1993). Finally, an impressive body of research indicates that self-reported and partner-reported maintenance behaviors are consistent and strong predictors of relational characteristics such as satisfaction, commitment, and trust (Dainton, Stafford, & Canary, 1994; Canary & Stafford, 1992, 1993; Stafford & Canary, 1991; Stafford et al. 2000; Dainton & Aylor, 2001; Afifi, Guerrero, & Egland, 1994; Bippus & Rollin, 2003; Fehr, 2000). Since relationship maintenance is defined as efforts to keep a relationship in a specified state or condition (Dindia & Canary, 1993, p. 164), most maintenance work has sought to link the performance of particular behaviors with relational qualities such as commitment, satisfaction, and uncertainty. Stafford et al. (2000) found that assurances, networks, and openness were positively correlated with satisfaction. The single best predictor of satisfaction and commitment is assurances, while openness and advice tend to be negative predictors of these same relational characteristics (Dainton & Aylor, 2002; Dainton et al., 1994; Stafford & Canary 1991; Stafford et al., 2000). Past scholars have also focused on investigating the relationship between maintenance and uncertainty. In Dainton and Aylor s (2001) study, moderate to strong, negative correlations were found between uncertainty and all five of Stafford and Canary s (1991) original maintenance strategies. Results from Ficara & Mongeau (2000) found negative associations between uncertainty and the use of assurances, openness, and positivity. Most recently, Dainton (2003) reported all seven maintenance strategies negatively correlated with uncertainty. Once researchers began to understand how maintenance functions in romantic relationships, scholarly interest in this topic shifted towards another common relationship friendships. 8

16 Friendship Maintenance Friendship maintenance and romantic maintenance are similar in at least three ways. First, paralleling Dainton and Stafford s (1993) research on romantic relationships, Emmett and Weger (2007) found that one cross-sex friend s use of a maintenance strategy predicts the other friend s use of the same strategy. This phenomenon also holds true in same-sex friendships (Oswald, Clark, and Kelly, 2004). Integration of these three studies indicates that maintenance behaviors correlate within dyads across several types of relationships. Second, as is the case in romantic relationships, some friendship maintenance behaviors, (i.e. shared activity, self-disclosure, and supportiveness), appear to be fundamental factors associated with emotional closeness and relational satisfaction (Afifi et al., 1994; Bippus & Rollin, 2003; Fehr, 2000): the chance of sustaining a comfortable and satisfying interpersonal relationship is potentially enhanced by engaging in positive relational maintenance behaviors. Finally, similarly to research on romantic relationships, Guerrero & Chavez (2005) found a strong, negative correlation between uncertainty and maintenance behaviors in cross-sex friendships. One explanation for this finding is that maintenance behaviors potentially act as a mechanism to reduce an individual s uncertainty about the relationship. Taken together, these three consistent patterns across relationship types may indicate uniform maintenance functioning in romantic relationships and friendships, however these few similarities lie within a larger pattern of dissimilarity; the remainder of this section reviews how maintenance patterns across friendship types differ. Results from Canary, Stafford, Hause, and Wallace s (1993) study indicate that people reported using many of the same maintenance behaviors with friends, romantic partners, and family members, however, respondents report less maintenance overall in friendships. Conclusions from 9

17 this study indicate that people put less emphasis on maintaining their friendships than their romantic and family relationships (Wiseman, 1986). Scholars (e.g., Guerrero & Chavez, 2005; Emmett & Weger, 2007) have established that the desire to escalate to a romantic relationship is associated with increased maintenance use. These findings complement Fehr s (1996, 2000) research; suggesting friends take one another for granted when there exists little interest in escalating the relationship. Wiseman (1986) concludes that less societal pressure exists for people to maintain or repair their friendships, which might explain why people spend less time maintaining these relationships. Since cross-sex friendships present their own special challenges (O Meara, 1989), people tend to maintain them differently than their other relationships. More specifically, Messman et al. (2000) identified six maintenance strategies for cross-sex friends. Positivity and openness paralleled the maintenance strategies found by Stafford and Canary (1991). Support (offering advice and comforting each other), avoidance of flirting (discouraging flirtatious or overly familiar behavior), sharing activities (spending time together or sharing special routines), and avoidance (acting in ways that prevent the friend from attaching romantically) emerged as similar, but distinct, categories of maintenance behaviors for cross-sex friends. These behaviors differ from romantic maintenance behaviors by discouraging rather than encouraging romantic feelings and commitment to the relationship. Further, Messman et al. (2000) found that cross-sex friends most heavily relied on support to maintain their friendships. Guerrero and Chavez s (2005) exploration of maintenance strategies across friendship types with respect to romantic intent concluded that additional categories of maintenance behaviors specific to cross-sex friendships exist, including relationship talk (in which friends talk directly about their friendship with each other), talk about outside romance (in which friends talk about their romantic relationships), avoidance of negativity (in which friends 10

18 avoid conflict and criticism of each other), and humor/gossip (in which friends joke with each other and share gossip). Not only do cross-sex friendships differ to some degree of maintenance when compared to other relationship types (Canary et al., 1993; Rose 1985), cross-sex friendships systematically differ from each other in terms of how people maintain them. Although relatively little research examined differences among cross-sex friendships in terms of their maintenance, two studies (Guerrero & Chavez, 2005; Emmett & Weger, 2007) investigated relational uncertainty and maintenance in crosssex friendships across levels of romantic intent (i.e., desire to remain strictly platonic, desire for romance, one partner desires romance, or one partner rejects romance). Results indicate that people s use of maintenance behaviors vary systematically with their romantic intentions (Guerrero & Chavez, 2005; Emmett & Weger, 2007). For example, in both studies, friends who mutually desired a romance engaged in the most maintenance behaviors in general. If at least one friend wants to escalate the friendship into a romantic relationship, it is clear that attraction exerts some influence on the relationship s trajectory. For this reason, the focus on this paper now turns to attraction within these female-male relationships. Attraction in Cross-Sex Friendships While past research shows that most male-female friendships do not involve sexual tension (Monsour et al., 1994), Swain (1992) reports that both sexes experience ambiguity about the sexual boundaries in some of these relationships. This ambiguity may stem from the fact that multiple types of emotional bonds and attraction types exist between men and women (Rawlins, 1982; Reeder, 2000). Although research on the role of sexuality in opposite-sex friendships is inconsistent, 11

19 Monsour et al. (1994) reports that sexual attraction and ambiguity give rise to potential costs in opposite-sex friendships. This finding complements Afifi and Burgoon s (1998) conclusions that ambiguity about the potential romantic/sexual nature of these friendships typically complicates relational functioning. Two of the four challenges O Meara (1989) identifies directly stem from sexual attraction. First, a friendship between a heterosexual man and woman introduces the potential for sexual attraction into the relationship. In addition, friends must use strategic communication to show relevant audiences they do not have a hidden sexual agenda. Consequently, cross-sex friends must mutually negotiate the boundaries of their relationship with each other and with outside parties. Since these challenges continuously exist throughout the lifespan of the friendship and never fully resolve themselves (O Meara, 1989), cross-sex friends must expend continued effort to ensure that they and others are clear about the sexual or nonsexual nature of the friendship; however, these tasks are not easy feats to accomplish. To review, Reeder (2000) identifies the following types of attraction in cross-sex friendships: subjective physical/sexual attraction (feeling physically or sexually attracted to the other person), objective physical/sexual attraction (thinking the other person is attractive in general, but not to oneself), romantic attraction (wanting to turn the friendship into a romantic relationship), and friendship attraction (feeling close and connected as friends). Also previously stated, scholars (i.e. Guerrero & Chavez, 2005; Emmett & Weger, 2007) have established that one type of attraction, the desire to escalate to a romantic relationship, is associated with increased maintenance use. This finding supports the idea that individuals in cross-sex friendships increase positive maintenance use when they prepare to escalate the intimacy level of 12

20 and/or the commitment level to the relationship. Overall, it seems clear from previous research that romantic attraction plays an important role in the way people maintain their cross-sex friendships. But, what happens though when attraction is put into action? As a starting point to answer this question, a review of friends-with-benefits research is presented below. Sex in Cross-Sex Friendships Young adults are having sexual relationships outside of the boundaries of romantic relationships, and are more often than not, choosing to do so with people that they know (Williams, Shaw, Mongeau, Knight, & Ramirez, 2007, p. 2). This may be one contributing factor as to why cross-sex friendships do not always remain strictly platonic. In fact, recent scholarly investigation (see Hughes, Morrison, & Asada, 2005; Mongeau, Ramirez, & Vorrell, 2003; Williams et al., 2007) suggests that some cross-sex friends actually seek out sexual contact with each other. Within the communication literature (and consistent with participants own vocabulary), this new emerging relationship is known as a friends with benefits relationship (FWBR). Hughes et al. (2005) defines FWBRs as relationships between cross-sex friends in which the friends engage in sexual activity but do not define their relationship as romantic (p. 1745). Mongeau et al. (2003) offer a similar definition: relationships in which friends in an otherwise platonic friendship engage in some degree of sexual intimacy on repeated occasions (p.1). Based on these definitions, ambiguity about the exact benefits in FWBRs exists. Since FWBRs vary in their degree of sexual intimacy, the benefits range from kissing up to, but not necessarily including, sexual intercourse. 13

21 While the actual benefits within FWBRs differ, research on the topic identifies three common factors about the nature of these relationships. Similar to romantic relationships, FWBR partners are chosen systematically. Typically, they are friends, former romantic partners, and/or desirable future partners (Williams et al., 2007). First, it is important to point out that sexual contact occurs between friends, not strangers or acquaintances (Mongeau et al., 2003). Given that these relationships are called friends with benefits, this might appear to be obvious. However, a friendship orientation to a non-romantic, sexual partner is what distinguishes a FWBR from a hookup. FWBRs differ from hookups, first, because hookups occur between strangers or brief acquaintances while FWBRs occurs between previously acquainted friends (Mongeau et al., 2003, p. 5). This shifts our attention to the second factor. The sexual contact is not just a one time occurrence within FWBRs. Instead, cross-sex friends participate in these relationships for repeated sexual contact within a non-committed, non-exclusive relationship (Williams et al., 2007). Finally, both members within FWBRs expect future interaction with one another. Again, this helps distinguish a hookup from a FWBR. As with most new research topics, early work on FWBRs (i.e., Hughes et al., 2005; Mongeau et al., 2003) sought to define the relationship context and primarily focused on the similarities between FWBRs. However, Williams et al. (2007) wanted to see how FWBRs differ from each other. Their results indicate that there are five different types of FWBRs. Just sex is the most distant form of a FWBR. Partners interact almost solely for sexual contact. Network opportunism seems to operate under the assumption if nothing better comes along. In this relationship type, sexual contact is initiated only when neither partner has found someone else for the night. The third type of FWBRs, labeled true friends with benefits most closely aligns 14

22 with existing scholarly definitions. In this case, friends who love, trust, respect, and consider the friend to be an important part of each other s lives engage in sexual activity. While these relationships are very similar to romantic relationships, they are not labeled in this manner by either of the partners. The last two types of FWBRs touch upon the fact that cross-sex friendships can be a trajectory towards establishing a romantic relationship and can also offer ex romantic partners a new relational context for continued interaction following a break-up. Transition in is a FWBR that intentionally serves to escalate the friendship into a committed romantic relationship. Here, cross-sex friends engage in sexual contact as a way to test the waters to see if they are sexually compatible with each other. The fifth and final type of FWBR is transition out. When romantic partners end their relationship, some couples continue to engage in sexual activity despite the fact they have broken up. In reviewing this research, it is clear that each type of FWBR produces different social and relational outcomes. In theory, FWBRs can end without damaging the participants egos and/or friendship (Denizet-Lewis, 2004), however, relational partners do not always mutually agree about the nature of their relationship (Williams et al., 2007). Men and women in these relationships may disagree about the nature of their past interaction, expectations for future interaction, desires for a romantic relationship, and levels of emotional investment. This significantly complicates relational functioning, which make relational maintenance patterns even more critical in this specific context. Just as romantic partners need to negotiate the seriousness or casualness of their relationships, FWBRs also need to negotiate the nature of their relationship. As one respondent from Williams et al. s (2007) study points out: [FWB is] the simple idea where two friends who do not date engage in sexual activities of some kind, yet it turns out to be very complex, not 15

23 simple (p. 5). Although respondents reported many reasons for this difficulty, the predominant reason was the one or both partners developed real feelings for the other. If this is the case, it is likely that their patterns of relational maintenance would align more closely with romantic maintenance than friendship maintenance. Although researchers have recently begun to study how sex operates within cross-sex friendships, no study has yet investigated sex s direct influence on friendship maintenance. Considering that 61.7% of college students sampled at one college campus had experienced relationships defined as FWBRs (Mongeau et al., 2003) and that these relationships are becoming increasingly common (Afifi & Faulkner, 2000; Manning, Giordano, & Longmore, 2006), a serious need exists for scholars to focus on this issue. It seems likely that relational maintenance behaviors will vary depending on whether or not the friends share a sexual history with each other. If cross-sex friends have engaged in past sexual contact with each other, they may maintain their relationship more similarly to a romantic relationship than a friendship. Since research (e.g. Canary et al., 1993; Wiseman, 1986) shows that maintenance is performed more in romantic relationships than friendships, cross-sex friends who have become sexually involved in the past may reasonably perform more maintenance than cross-sex friends who have remained platonic throughout their friendship s duration. Furthermore, one possible reason why cross-sex friends may engage in sexual contact with each other is to signify a desire to escalate the friendship to a romantic relationship. Since scholars (Guerrero & Chavez, 2005; Emmett & Weger, 2007) have found that this desire is associated with increased maintenance use, this supports the assumption that cross-sex friends who have been sexually involved in the past may 16

24 maintain their relationships more than their non-sexually involved counterparts. For these reasons, the researcher proposed the following hypothesis: H1: Higher levels of relational maintenance will be performed by cross-sex friends who have been sexually involved in the past than by cross-sex friends who have shared a strictly platonic relationship. Although sexual contact between cross-sex friends may increase relational maintenance by increasing the friendship s baseline intimacy level, sexual experiences are not always perceived positively. Sexual contact may be a way for cross-sex friends to test the waters to see if they are compatible on a romantic level. If sexual activity was positively perceived, patterns of maintenance may start to resemble romantic relationships. However, negative perceptions of sexual contact may prohibit the use of pro-social maintenance behaviors. It seems likely that positive experiences of sex would be more rewarding to relational functioning and maintenance than negative experiences; however, this relationship has never been explored. Thus, the researcher proposed the following question: RQ1: Does the valence of past sexual experience affect present patterns of relational maintenance within cross-sex friendships? Dating Status One factor that has gone unstudied when considering how attraction operates on maintenance is the dating status of both cross-sex friends. While past studies methods sections include information about the participants dating status, past researchers refrain, however, from including this as a variable in data analysis. This methodological shortcoming restricts scholars understanding of how attraction influences communication patterns. In using a method that 17

25 incorporates dating status as an independent variable, researchers can better understand the direct effects of symmetrical/asymmetrical dating status on maintenance use in cross-sex friendships. Considering that individuals in committed romantic relationships have less time to invest in their friendships in general, this appears to be an important variable of interest to researchers. In addition, a romantic partner potentially becomes jealous if his or her significant other spends time with another person, especially if this person could be a potential mate. Since expressions of intimacy through physical or sexual contact were significantly more common in cross-sex friendships than in same same-sex friendships (Monsour, 1992), same-sex friends are not often viewed as threatening to a romantic relationship as opposite-sex friendships. Consequently, when involved in a heterosexual, committed relationship, same-sex friendship maintenance is less taboo than opposite-sex friendship maintenance. Thus, individuals in committed relationships may strategically engage in less cross-sex friendship maintenance in consideration of their romantic partner s feelings of insecurity or jealousy. Further, since many men and women establish friendships with one another as a trajectory step towards establishing a romantic relationship, it is likely that individuals already in romantic relationships are less motivated to maintain opposite-sex friendships. All of these reasons suggest a main effect for dating status on maintenance. Thus, the researcher proposed the following hypothesis: H2: The lowest levels of maintenance will exist in relationships where both friends are in committed relationships. Ultimately, this exploratory work will deepen our scholarly understanding of relational maintenance processes specifically within cross-sex friendships. 18

26 CHAPTER 3: METHODS Participants Undergraduate students were recruited from lower and upper division communication courses at a large southeastern university. Of the 297 total students who completed questionnaires, 20 who identified themselves or their friend as homosexual, unsure about their sexual orientation, or preferred not to report on their sexual orientation were excluded from this study because the issue of romantic attraction would be different for these participants than for those pairs of friends where both were identified as heterosexual or bi-sexual. Self-report data from the participants show that 272 are heterosexual and five are bi-sexual. When participants reported on their cross-sex friend s sexual orientation, 265 of the friends were identified as straight and 12 as bisexual. The average age of these 277 respondents (female n = 159, male n = 118) was 19.8 years old (range = years old; md = 19 years old). Of the sample, 24.5% were freshman, 37.5% were sophomore, 27.1% were juniors, 10.5% were seniors, and 0.40% were graduate students. With respect to ethnicity, the sample was 67.9% Caucasian, 17.3% Hispanic, 7.2% African-American, 4.7% Asian, and 2.9% identified themselves as other. Respondents had known their cross-sex friends for an average of 43.3 months (range = months). Regarding friendship level, 56.0% of the sample reported on a close friend, 21.3% reported on a best friend, and the remaining 22.7% reported on a casual friend. At the time of the study, 49.8% of the respondents were single, 9.0% were casually dating, 38.3% were in committed relationship, 2.5% were engaged, and 0.40% were married. None of respondents reported that they were divorced or widowed. 19

27 When reporting on their cross-sex friend s dating status, respondent reported that 54.2% were single, 21.7% were casually dating, 19.9% were in a committed relationship, 1.1% was engaged, 2.5% were married, and 0.70% was divorced or widowed. Of the sample, 26.7% of respondents reported engaged in sexual contact with their cross-sex friend in the past, 72.6% reported no history of sexual contact with their cross-sex friend, and the remaining 0.4% preferred not to respond to this question. Of the 74 participants that reported a sexual history with their friend, 23.4% reported it as a very positive experience, 37.7% reported it as a positive experience, 35.1% reported it as neither a positive nor negative experience, 1.3% reported it as a negative experience, and 2.6% reported it as a very negative experience. Procedures The researcher visited various undergraduate communication classes to recruit volunteers, who were informed that the questionnaire examined a wide range of behaviors that people use in friendships between men and women. Questionnaires were distributed and completed in both large lecture and smaller discussion format classes by students who had a cross-sex friend and were above the age of 18. The questionnaire (see Appendix C) consisted of a self-report measure of relational maintenance behaviors, attraction type, dating status, relationship satisfaction, friendship level, sexual/romantic history, and future sexual/romantic intentions. Students were told not to place their names on the questionnaire to ensure anonymity. instructions: Instrumentation On the first page of the survey, respondents were supplied with the following set of 20

28 This questionnaire involves recalling and reporting the activities you use in a friendship with a cross-sex friend. Cross-sex friends are defined as friends of the opposite sex whom you spend time with, but do not currently date. Please think of a friend fitting this description and keep this friend in mind as you answer all of the following questions. At the beginning of each subsequent section of the survey, the participant was reminded to keep this particular person in mind as they responded to the set of questions. All survey items were recoded so that higher scores reflect higher levels of the construct under investigation. Relationship Maintenance Behaviors To obtain a comprehensive measure of relationship maintenance behaviors, two instruments developed by Emmett and Weger (2007) and Oswald, Clark, and Kelley (2004) were combined and included on the survey to operationalize relational maintenance. Together, these scales included 74 Likert-type items, asking participants how frequently they do each particular behavior (1 = Never; 2 = Rarely; 3 = Sometimes; 4 = Often; 5 = Always). A principal component factorial analysis was conducted, using varimax rotation. Based on eigen values over 1.0 and interpretability, 10 different sets of relationship maintenance behavior subscales were derived. From the original 74 items, 56 items were retained. All of these subscales were consistent with those previously found in research (see Guerrero & Chavez, 2005; Emmett & Weger, 2007; Stafford & Canary, 1991). Table 1 identifies all of the items and the subscale s primary loading, however, a brief overview of the subscales is presented here. Support availability is a combination of instrumental and emotional support. These maintenance behaviors measure the degree to which friends make themselves available to help in with everyday coping. Routine contact and activity is the collective group of behaviors that friends make to spend time with or contact each other on a consistent basis. Relationship 21

29 talk is the effort friends make to talk directly about their friendship with each other. Avoidance of negativity is the collective group of behaviors that friends make to avoid conflict and criticism of each other. Positivity are the behaviors that creative positive and enjoyable interactions for friends. Social networking is the effort friends make to spend time with common friends and affiliation. Flirtation is encouraging flirtatious and overly familiar affection. Indirect communication is the degree friends use mutual third parties to express or inquire about the status of their friendship. Talk about outside romance is the effort friends make to talk about their current/past romantic relationships. Lastly, Humor is when friends joke with each other. While most of the subscales met conventional levels of reliability (Support availability =.92; Routine contact and activity =.88; Relationship talk =.89; Avoidance of negativity =.77; Positivity =.85; Social networking =.81; Flirting =.80; Indirect communication =.75, Talk about outside romance =.88) one of the subscales had slightly below acceptable levels of reliability (Humor =.65). Since this subscale consisted of only two items, and given that subscales with three or fewer items typically produce lower levels of reliability, and considering this subscale s reliability is only slightly less than the conventional norm, it was retained in the analysis. Descriptive statistics for each maintenance variable are depicted in Table 2. Sexual/romantic history This variable was measured with the following two questions: Have you ever been in a romantic/dating relationship with your friend in the past? and Have you ever hooked up/had any sexual contact/had sex with your friend in the past? Friends who answered no to both of these questions were said to have no sexual involvement in the past. For data analysis purposes, 22

30 this variable was dummy coded 0 for no sexual involvement in the past, and 1 for sexual involvement in the past. Dating status When collecting demographic information at the beginning of the survey, respondents reported both their and their friend s dating/marital status. Both response sets included single, casually dating, in a committed relationship, engaged, married, or divorced/widowed. For data analysis purposes, these five categories were collapsed into two levels: single and taken. The Single category included the Single, Casually Dating, and Divorced/Widowed responses. The Taken category included the Committed Relationship, Engaged, and Married responses. Since the second hypothesis concerns combinations of the friend s and the participant s own dating status (i.e., comparisons between both involved in an outside relationship or both single), Rosnow and Rosenthal s (1996) approach was followed by creating a single factor from the four possible cells of the Friend s Status by Own Status factorial. In other words, all possible combinations of self by other dating status were recoded into four categories (i.e., a 1 X 4 factorial): Friend Single/ Participant Single, Friend Single/Participant Taken, Friend Taken/ Participant Single, and Friend Taken/ Participant Taken. Friendship level While not of empirical interest in this analysis, friendship level was included as a way to control extraneous source of variance in relational maintenance. Oswald et al. (2004) indicates that best friends engage in more maintenance than close friends, and close friends engage in more maintenance than casual friends. In the demographics section of the survey, respondents 23

31 were asked How would you classify your friend? Consistent with Johnson, Wittenberg, Villagran, Mazur, and Villagran (2003), friendship levels in this study were classified as casual, close, or best. 24

32 CHAPTER 4: RESULTS Since the independent variables (past sexual involvement, dating status, friendship level) are all categorical data and the dependent variables (relational maintenance behaviors) are all interval-level data, a single 1X4 (combined dating status) MANCOVA with friendship level and past sexual encounter as covariates was performed to determine the effect of the independent variables on the relationship maintenance variables as a whole. Friendship level was entered as a control variable. Consistent with previous research MANCOVA results indicate friendship level does predict overall maintenance behaviors with best friends performing more overall maintenance than close or casual friends (See Table 3 for MANCOVA statistics, probability levels and effect sizes). An examination of the parameter estimates (see Table 4) indicate that increasing friendship level is associated with higher levels of six of the ten maintenance behaviors, including support availability routine contact and activity, relationship talk social networking), talk about outside romance and humor Although these findings are not directly related to the purpose of this study, they support the decision to use friendship level as a control variable for the rest of data analysis. Hypothesis One Hypothesis 1 predicted that cross-sex friends who engaged in past sexual contact would perform more maintenance than cross-sex friends who never engaged in sexual contact. As indicated by the MANCOVA s results (see Table 3), overall maintenance levels do differ as a function of past sexual involvement. Since the MANCOVA test was significant, parameter estimates can be interpreted to determine which specific maintenance behaviors are affected by 25

33 this variable. As reported in Table 4, past sexual involvement was associated with higher levels of three of the ten maintenance behaviors: relationship talk (B =.24, p =.02), avoidance of negativity (B = -.16, p =.04), and flirting (B =.81, p <.001). In sum, H1 was partially supported. Research Question One Research Question 1 was examined by regressing Valence of Past Sexual Encounters on Friendship Level and all of the maintenance variables. This backwards regression strategy helps to identify the relationships among Valence of Past Sexual Encounters and the relationship maintenance strategies by producing partial correlation coefficients between each maintenance variable and the dependent variable. The initial regression model entered Friendship Level on the first step to control for extraneous variance and then the balance of the maintenance variables on the second step. The resulting model fit the data well, R 2 =.33, p =.005. However, two maintenance variables, Support Availability and Routine Contact and Activity produced high variance inflation (VIFs = 2.71 and 2.38 respectively). Variance inflation increases Type II error, in this case more than doubling the error variance, making significance tests too conservative. A comparison of two regression models (model 1 including, and model 2 excluding, Support Availability and Routine Contact and Activity) indicated that the inclusion of Support Availability and Routine Contact and Activity did not significantly increase the fit of the regression model, R 2 change =.01, F change =.551, p =.58. Therefore, a second model was computed without Support Availability and Routine Contact and Activity. The second model produced a good overall fit with the data, R 2 =.31, F = 3.29, p =.002. Analysis of the partial correlation 26

34 coefficients suggests two maintenance variables, Positivity and Flirting, along with Friendship Level, display significant positive associations with Valence of Past Sexual Encounters (See Table 7 for details). Hypothesis Two Hypothesis 2 predicted that the lowest levels of maintenance would exist in the cross-sex friendships in which both friends are in committed relationships. MANCOVA results (as shown in Table 3) indicate that the combination of relational maintenance variables differ over levels of own/friends dating status. The means and standard deviations for each maintenance variable across levels of dating status are presented in Table 5. Since the MANCOVA was significant, step down F-tests were performed to identify and determine how specific maintenance behaviors were affected by the four combinations of dating statuses. Results (see Table 6) indicate that dating status significantly affects 3 of the 10 maintenance behaviors: routine contact and activity, flirting, and humor. Contrast tests were then computed to determine the nature of the differences among dating status combinations. As shown in Table 6, a linear relationship was found with respect to routine contact and activity, indicating that the most routine contact and activity was performed by the both single friendships and the least was performed by the both taken friendships (see Figure 1). For flirting, a quadratic relationship was found indicating that the friendship types both single and both taken flirted the most and the single/taken friendship types flirted the least (see Figure 2). Finally, a cubic relationship was found with respect to humor indicating a complex relationship in which it appears that both taken and both single relationships are some place in the middle with friend taken/participant single friendships 27

35 engaging in the least humor and friend single/participant taken engaging in the most humor (see Figure 3). 28

36 CHAPTER 5: DISCUSSION If communication scholars could be considered parents of the subjects they research, cross-sex friendships could be comparable to being the youngest child in that family unit. Until recently, cross-sex friendships have gone somewhat unnoticed on the research scene and have primarily received hand-me-down theories from romantic relationships. Since men and women today have greater opportunities to develop friendships with members of the opposite sex and past research has determined that these friendships differ than romantic relationships, this study sought to bring cross-sex friendships into the research foreground for a well-deserved, detailed analysis. This investigation sought to answer three specific questions about relational maintenance in cross-sex friendships. First, how does past sexual involvement between cross-sex friends affect their current maintenance patterns? Second, are present patterns of maintenance affected by how cross-sex friends perceive their experiences of past sex? And last, how does both friends dating status affect maintenance patterns? The remainder of this paper explains how the current study provides some answers to these questions and outlines how future researchers could contribute to developing more detailed responses to these same inquiries. With regard to the first question, this study concludes that past sexual involvement between cross-sex friends affects current patterns of friendship maintenance in at least three ways. Past sexual involvement was positively associated with three maintenance behaviors: relationship talk, avoidance of negativity, and flirting. Even though FWBRs are becoming increasingly common (Mongeau et al., 2003; William et al., 2007; Afifi & Faulkner, 2000; 29

37 Manning, Giordano, & Longmore, 2006), socialization patterns establish that sexual contact is usually reserved for romantic relationships. Given that our culture chiefly emphasizes the normalcy of sexuality in a romantic context, cross-sex friends who had sexual contact in a nonromantic relationship may find it necessary to talk more about their relationship status. A crosssex friendship sometimes can be used as a trajectory towards a romantic relationship; however, this is not always the case. Therefore, talking about their friendship may help cross-sex friends reduce uncertainty about the current/future nature of their relationship. In addition, via talking about the relationship they may be able to understand each other s motivation to have sex. Next, with respect to avoidance of negativity, results suggest that friends who never crossed the line are more likely to engage in negativity with each other (i.e., ignore or blame one another, threaten to end the friendship, or talk badly about the other). Cross-sex friends who share a sexual history with one another would probably have more difficulty classifying their relationship as just a friendship than cross-sex friends who have never been sexually involved. Since research shows less pressure exists for individuals to maintain friendships than romantic relationships, cross-sex friends with no sexual history probably feel less pressure to maintain their relationships than their once sexually active counterparts. Thus, cross-sex friends who have never become sexually involved may consequently experience more freedom to take risks in maintaining these friendships, which helps explain why they may engage in more negative behaviors. Last, higher levels of flirting were associated with past sexual involvement. Quite simply, high levels of flirting may have been an antecedent to or consequence of the cross-sex friends sexual activity. Increased amounts of flirting could lead to sexual contact or sexual contact could 30

38 motivate friends to flirt more. Indeed, if these are friends with benefits relationships, flirtation ought to be expected as a routine activity in probing for interest in future sexual encounters. RQ1 asked if positive perceptions of sexual history between cross-sex friends influence maintenance differently than negative ones. Data analysis suggests two maintenance variables, Positivity and Flirting, along with Friendship Level, display significant positive associations with valence of past sexual encounters. Although this study is primarily focused on maintenance behaviors, it is important to note that Friendship Level was a significant factor in the equation model. Results indicate that as friendship level increases, the valence of sexuality activity is perceived more positively. One explanation may be that increased levels of friendship are associated with higher levels of intimacy and feelings of comfort and lower levels of performance anxiety. With respect to maintenance behaviors, the valence of sexual activity influenced how often positivity and flirting occurred. Participants who reported less satisfying sexual experiences with a cross-sex friend engaged in less positivity with their friend. In other words, the more satisfied respondents perceived their sexual activity, the more positively they interacted with their friends. Although causality can not be determined, perceptions of satisfying sexual experiences may lead to increased levels of positivity or positive interactions since the satisfying sexual experience may create feelings of intimacy and warmth toward the friend. It is also possible that a highly positive climate within the friendship may also lead to a more satisfying sexual experience. The possibility also exists that this relationship is reciprocal so that the sexual experience influences positivity and the resulting increases in positivity result in satisfying sexual encounters in the future. 31

39 With respect to the second maintenance strategy, the more positively the sexual encounter was perceived, more flirting occurred in these friendships. Cross-sex friends who negatively perceive their past sexual encounters flirt less than their more sexually satisfied counterparts. Again, directionality of the causal relationships can not be clearly established, but two possibilities are explored here. It may be that increased levels of flirting function as communicative fore-play, which is conducive to future sexual encounters. Another alternative is that positive sexual experiences lead to increased levels of flirting. In this case, the act of flirting can function as indirect offers for future sexual activity. While flirting could be an indirect means to test romantic compatibility, perceptions of failed sexual contact may directly discourage future flirting attempts. Finally, it is worthwhile to note that Relationship Talk approached significance levels. The more positively sex was perceived, the less friends talked about their relationship. It may be that cross-sex friends are more willing to accept ambiguity about their relationships status when they are satisfied with their sexual experiences. H2 predicted that the lowest levels of maintenance would exist in relationships where both friends are in committed relationships. Results indicate that overall maintenance patterns differ by friends dating statuses. Specifically, amounts of routine contact and activity, flirting, and humor systematically differed according to the participants and their friend s dating status. Importantly, this study demonstrates that the relationships between maintenance behaviors and being in a committed relationship are not only linear ones. While a linear relationship was found with respect to routine contact and activity, a quadratic relationship was found with respect to flirting, and a cubic relationship was found with respect to humor. It is important to evaluate 32

40 what each of these relationship types can help scholars understand about cross-sex friendships dynamics. When both friends are in committed relationships, routine contact and activity is at its lowest. Serious romantic relationships significantly reduce the amount of time and energy friends can spend with or use to communicate with each other. If maintenance demands increase and personal resources remain constant, it is probable that a person will focus more energy on the relationship he/she feels more pressure to maintain. In light of the fact that romantic relationships take priority in our country, this result seems intuitive. It is probable that there is a shift in maintenance efforts. Some of the previous energy he/she expended on the friendship may now shift towards his/her current romantic relationship. Flirting was also found to differ across dating status, however, this relationship is less straightforward. Almost equivalent high amounts of flirting occurred when both friends were single and when both friends were taken. These diverging results may be explained by examining a person s motivation to flirt. The original predication that flirting would be highest when both friends were single was supported. If the motivation to flirt is to signal a romantic interest in the other person and the target is already taken, this communication could be considered failed attempts. Following this logic, higher levels of flirting would occur in friendships where both people are single than in friendships where only one person is single. The results acknowledge this possibility, while simultaneously allowing for another scenario. If the motivation to flirt is simply to engage in fun, light-hearted interaction with someone of the opposite sex, flirting in cross-sex friendships could function in this manner if both friends understood this innocent motivation. If both friends are in committed relationships with outside 33

41 parties, flirting between cross-sex friends could become romantically meaningless and viewed simply as a preferred way of communicating. Humor was the third variable to vary across dating status. Approximately equal amounts of humor were used by participants when both they and their friend were single and were both taken. However, results show that levels significantly differed when the friends were not in the same category. Lowest levels of humor occurred when the respondent was single and the friend was taken; highest levels of humor occurred when the respondent was taken and the friend was single. Once again, these results may be explained by examining the function of humor in crosssex friendships. Since humorous interaction between members of the opposite sex could be misconstrued as flirting or could provoke physical affection, single cross-sex friends may reduce this type of interaction with their taken friends to avoid sending mixed signals to the person involved in the romantic relationship. Humor can often be misinterpreted. If the single friend uses less humor, he/she may produce less confusing situations for the taken friend. However, once the respondent becomes involved in a committed relationship, results indicate that he/she may again feel free to humorously interact with their friends of the opposite sex. Limitations and Future Research Despite its limitations, this study contributes to scholarly understanding of cross-sex friendships and relational maintenance processes. Future researchers are encouraged to extend this line of research by pursuing unexplored areas of interest, some of which are offered here. This study s most obvious limitation is the lack of generalizability that comes from using a convenience sample of college students. Although a respectable sample size was collected, it is 34

42 recommended to collect data from an older sample. This would enable researchers to investigate with more detail how cross-sex friendships are maintained into married life and beyond. Also, considering that sexual activity outside of a romantic context may be considered taboo behavior, participants may have been preoccupied with social desirability when reporting on their sexual history with their cross-sex friend. It is recommended that researchers consider using a qualitative approach for future studies. In doing so, scholars could begin exploring what each relational maintenance behavior actually means to the people who use them in their friendships. Another area of concern that is important to address is how scholars operationalize crosssex friendships on their survey instruments. Without question, a researcher s methods affect what he/she finds. Respondents were informed that cross-sex friends are defined as friends of the opposite sex whom you spend time with, but do not currently date. Future researchers should consider changing the wording of these directions to exclude the word friend. If researchers are interested in studying relations between men and women where sexual tension presently exists or past sexual behavior has occurred, researchers should refrain from labeling these relationships as friendships. When asked to report on a cross-sex friend, participants probably will report on relationships that most closely resemble the prototypical friendship. Directions for future survey should be re-worded to address this problematic concern. This study focused on how patterns of relational maintenance were affected by past sexual attraction and activity. Given that this sample had a relatively long average of friendship length, it seems probable that cross-sex friends who were going to have sexual contact with each other would already have done so. However, it is important to acknowledge that sexual tension and attraction is not always acted upon. Scholarly understanding of maintenance processes could 35

43 be expanded if researchers examined how latent sexual tension affects maintenance functioning. Scholars now can look at how future intentions for sexual involvement may affect present patterns of relational maintenance. Although McCrosky and McCain s (1974) claim that, interpersonal attraction does appear to be a multidimensional construct (p. 266.), studies on cross-sex friendships typically concentrate on sexual attraction as the sole focus of research. Although this signifies our society s general assumption of sexual attraction s prevalence in male-female friendships, this works against expanding our academic understanding of interpersonal attraction, in general. Future researchers would profit from investigating how other forms of attraction manifest and affect the performance of maintenance work. From a practical perspective, when cross-sex friends cross the line by becoming either involved with each other or involved in a committed relationship with others, both friends should not perceive decreases in maintenance as an insult to the friendship. Overall, this study indicates that patterns of relational maintenance are expected to change throughout the course of a friendship as sexuality becomes salient and romantic relationships develop or dissolve. If friends can learn to see these ups and downs as naturally occurring processes in their relationship, they can actually learn to become better friends to each other. 36

44 APPENDIX A: IRB APPROVAL LETTER 37

45 38